Oil and gas pipeline threat event identification method and system

Abstract

The invention discloses an oil and gas pipeline threat event identification method and system, and relates to the technical field of oil and gas pipeline integrity management and oil and gas pipeline protection, and the method comprises the following steps: S1, building a digital elevation model of a target area; s2, acquiring position attitude data of the unmanned aerial vehicle; s3, acquiring a real-time video stream of the unmanned aerial vehicle camera, and identifying a potential oil and gas pipeline threat event in the real-time video stream; s4, aligning the key video frame data of the potential oil and gas pipeline threat event with the position and attitude data of the unmanned aerial vehicle; s5, transforming the aligned camera coordinate projection of the key video frame of the threat event into a world coordinate by using a digital elevation model, and calculating a real space coordinate position corresponding to the video frame image of the threat event and the projection of a pipeline in the key video frame; and S6, displaying the threat event. According to the invention, the timeliness is improved by adopting a real-time analysis technology, and early warning and prompting can be carried out at the first time after a potential target is found.

Description

Technical Field

[0001] This invention relates to the fields of oil and gas pipeline integrity management and oil and gas pipeline protection technology, and more specifically to a method and system for identifying oil and gas pipeline threat events. Background Technology

[0002] With advancements in technology and increased automation, drones have become an efficient and safe solution for oil and gas pipeline inspection, playing a significant role in pipeline protection. Compared to traditional inspection methods, drone inspections significantly reduce manpower and time costs. Drones can quickly cover large pipeline networks, dramatically improving inspection efficiency. Equipped with high-resolution cameras and sensors, they provide high-precision inspection data. For inspections in complex terrain or hazardous areas, using drones greatly reduces safety risks for personnel.

[0003] Current technology typically involves manually exporting video and image files from the drone's camera after a pipeline inspection, then stitching the images together to generate a strip-shaped orthophoto of a certain area around the entire pipeline. Image recognition is then performed on the stitched image to identify specific pipeline threats. The problems with this method are that image stitching requires a certain level of technical skill from data processing personnel, the stitching process is lengthy, the generated strip-shaped orthophoto file is very large, it demands high-level computer image processing capabilities, and there is a lag in target identification. Summary of the Invention

[0004] To overcome the shortcomings of the existing technology, this invention discloses a method and system for identifying oil and gas pipeline threat events. This invention uses real-time analysis technology to improve timeliness and can provide early warning and alerts as soon as a potential target is discovered.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] In a first aspect, the present invention provides a method for identifying oil and gas pipeline threat events, comprising the following steps:

[0007] I. Establishment of Digital Elevation Model

[0008] S1. Establish a digital elevation model of the target area;

[0009] Preferably, step S1 includes: acquiring three-dimensional coordinate point cloud data of the target area using a laser scanning radar mounted on a UAV, filtering and classifying the three-dimensional coordinate point cloud data, and establishing a digital elevation model of the target area.

[0010] II. Acquisition of UAV position and attitude data

[0011] S2. Acquire the position and attitude data of the UAV;

[0012] Preferably, step S2 includes: mounting a computing processing load on the UAV, wherein the computing processing load acquires the UAV's position and attitude information at a frequency of 1Hz through interface programming, wherein the position and attitude information includes longitude, latitude, altitude, yaw angle, pitch angle and roll angle.

[0013] III. Identification of Potential Threats to Pipeline Security

[0014] S3. Acquire real-time video streams from drone cameras and identify potential oil and gas pipeline threat events in the real-time video streams;

[0015] Preferably, step S3 includes: acquiring the real-time video stream from the drone camera using a computational processing load; processing and analyzing the real-time video stream using a pipeline threat target identification model built within the computational processing load; identifying potential pipeline security threats in the video stream; and denoting the identified targets as E. i Record E i The coordinates of the top left corner (X) s Y s ) and the coordinates of the lower right corner (X) e Y e (and the corresponding timestamp.)

[0016] IV. Data Alignment

[0017] S4. Align key video frame data of potential oil and gas pipeline threat events with UAV position and attitude data;

[0018] Preferably, step S4 includes:

[0019] The computational processing load aligns the key video frame data of the identified potential threat targets with the UAV pose data acquired in step S2 using timestamps;

[0020] If the video frame timestamp and the pose timestamp cannot be perfectly aligned, then the two sets of data closest to the video frame timestamp in the pose information are used for interpolation.

[0021] Preferably, in step S4, the interpolation includes linear interpolation, which calculates the interpolation values ​​for height and attitude angle. For any attribute x, the attribute value between two time points t1 and t2 is:

[0022]

[0023] Where x(t) is the attribute value at time t, x(t1) and x(t2) are the attribute values ​​at two time points, t is the time point to be interpolated, t1 is the time point at which interpolation begins, and t2 is the time point at which interpolation ends.

[0024] Preferably, in step S4, the interpolation includes spherical linear interpolation, which calculates longitude and latitude, including:

[0025] △λ=λ2-λ1

[0026] a=sin((1-f)×Δλ)×sin(λ1)+sin(f×Δλ)×sin(λ2)

[0027] b=sin((1-f)×Δλ)×cos(λ1)+sin(f×Δλ)×cos(λ2)

[0028] λ(t) = atan2(a,b)

[0029] Where λ1 and λ2 are the longitude or latitude of the start and end time points, respectively; Δλ is the difference between the two coordinates; and f is the proportion from the start time point to the interpolation time point, calculated as follows: t is the time point to be interpolated, t1 and t2 are the time points being interpolated, a is an intermediate value calculated based on the longitude / latitude of the starting and ending points and the interpolation factor f, taking into account the changes in longitude / latitude and the sine value of the starting point's longitude / latitude, b is another intermediate value, combining the cosine values ​​of the starting and ending points' longitude / latitude and the sine value of the changes in longitude / latitude, and λ(t) is the longitude / latitude at time t.

[0030] V. Projection Transformation

[0031] S5. Using a digital elevation model, the camera coordinates of the aligned key video frames of the threat event are projected and transformed to world coordinates to calculate the real spatial coordinates of the threat event video frame image and the projection of the pipe in the key video frame.

[0032] Preferably, step S5 includes: using a camera coordinate to world coordinate conversion method to convert the coordinates of the potential threat event to obtain the rectangular spatial range of the threat event in world coordinates.

[0033] Preferably, step S5 includes: using a spatial computation method to calculate each threat event E. i The distance to the pipeline is used to determine the type of potential threat event and to label the threat event.

[0034] Preferably, step S5 includes:

[0035] By using a camera coordinate to world coordinate conversion method, the geographic spatial location corresponding to the entire image of the video key frame is calculated. Spatial search analysis is performed by comparing the geographic spatial range corresponding to the image with the spatial coordinates of the actual pipeline route to find the pipeline route coordinates falling in the image.

[0036] By using the transformation method from world coordinates to camera coordinates, the pipeline route is superimposed with key video frames to express the relative positional relationship between potential threat events and pipelines on the image.

[0037] Preferably, in step S5, the method for converting camera coordinates to world coordinates includes: converting camera coordinates (x...) to world coordinates. c ,y c 1) Transform to world coordinates (X) by inverse rotation and translation. w ,Y w Z w ),in:

[0038] Ray direction:

[0039] Ray origin: o = R -1 ×(-T)

[0040] Intersection of the ray and the ground: P w =o+t×d

[0041] Where R is the rotation matrix and T is the translation vector (Ti). x ,T y ,T z ), that is, the latitude, longitude and altitude coordinates of the UAV obtained from step S2, t is the scaling factor of the ray from the starting point to the ground intersection point. The ray starts from the position of the camera and extends in a specific direction until it intersects the ground.

[0042]

[0043] Among them, Z ground It is the ground elevation, obtained from the digital elevation model established in step S1. z It is the Z-coordinate of the ray origin in world coordinates, d z It is the component of the ray direction vector on the Z-axis.

[0044] Preferably, in step S5, the transformation method from world coordinates to camera coordinates includes:

[0045] Point P in the world coordinate system w =[X w ,Y w Z w [Through rotation matrix R and translation vector T(T)] x ,T y ,T z Transform to camera coordinate system P c =R×P w +T;

[0046] Projective transformation from camera coordinates to image coordinates, point P in the camera coordinate systemc =[X c ,Y c Z c The image coordinates P are generated by projecting the camera intrinsic parameter matrix K onto the image plane. i =[u,v]:

[0047] Preferably, the camera intrinsic parameter matrix K is:

[0048]

[0049] c x =Image width / 2

[0050] c y =Image height / 2

[0051] The rotation matrix R is a rotation matrix R(yaw, pitch, roll) constructed from the rotation of each angle. z yaw), R y (pitch), R x (roll), the total rotation matrix is ​​the product of these matrices: R = R z (yaw)×R y (pitch)×R x (roll), where:

[0052]

[0053] Where roll represents the roll angle, pitch represents the pitch angle, and yaw represents the yaw angle.

[0054] Preferably, the transformation from image coordinates to camera coordinates includes: given image coordinates (u,v), transforming them to normalized camera coordinates (x,v). c ,y c The process includes:

[0055]

[0056] VI. Threat Event Display

[0057] S6. Display the threat events.

[0058] Preferably, step S6 includes: displaying the threat events and overlaying them with the results of multiple rounds of identification to determine which threat events are new and which are historically identified.

[0059] Based on the above-mentioned method for identifying oil and gas pipeline threat events, the present invention also provides an oil and gas pipeline threat event identification system, comprising:

[0060] Hardware modules include drones, LiDAR, cameras, and computing workloads;

[0061] The computational models include digital elevation models, position and attitude models, pipeline threat target identification models, data alignment models, projection transformation models, and overlay analysis models;

[0062] The data module includes hardware-acquired data and model-analyzed output data.

[0063] Preferably, the hardware acquisition data includes the vertical distance acquired by the lidar, the video stream and timestamp acquired by the camera, and the UAV flight control data such as longitude, latitude, altitude, yaw angle, pitch angle, roll angle, and coordinates of the threat target.

[0064] The beneficial effects of this invention are:

[0065] 1. This invention employs real-time video stream analysis technology from unmanned aerial vehicles (UAVs), using an embedded computing load to identify threat events in real time during UAV inspections. Compared to traditional image stitching and offline analysis methods, this significantly shortens the identification time and can trigger an early warning immediately upon detecting potential threat targets, thus significantly improving the efficiency of identifying oil and gas pipeline threat events.

[0066] 2. This invention, through the integrated application of a high-precision digital elevation model (DEM) and attitude data, accurately projects the coordinates of key video frames of potential threat events onto the world coordinate system. By combining spherical linear interpolation and spatial calculation methods, errors in data alignment and projection are effectively reduced, ensuring the accuracy of the identification results. This allows for a more precise determination of the type of threat event and its impact on the pipeline.

[0067] 3. This invention can identify various threat events in real time, such as third-party construction, encroachment, and insufficient spacing, and through spatial overlay analysis, it can identify both new and historical threat events. This comprehensive monitoring and analysis capability ensures the safety and integrity of pipelines in various environments, providing comprehensive and powerful technical support for the protection of oil and gas pipelines. Attached Figure Description

[0068] Figure 1 This is a schematic diagram of the oil and gas pipeline threat event identification method of the present invention. Detailed Implementation

[0069] The following will provide a clear and complete description of the concept, specific structure, and technical effects of the present invention in conjunction with the embodiments and accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention.

[0070] Example 1

[0071] A method for identifying oil and gas pipeline threat events, characterized by comprising the following steps:

[0072] S1. Establish a digital elevation model of the target area;

[0073] S2. Acquire the position and attitude data of the UAV;

[0074] S3. Acquire real-time video streams from drone cameras and identify potential oil and gas pipeline threat events in the real-time video streams;

[0075] S4. Align key video frame data of potential oil and gas pipeline threat events with UAV position and attitude data;

[0076] S5. Using a digital elevation model, the camera coordinates of the aligned key video frames of the threat event are projected and transformed to world coordinates to calculate the real spatial coordinates of the threat event video frame image and the projection of the pipe in the key video frame.

[0077] S6. Display the threat events.

[0078] Example 2

[0079] This embodiment further elaborates on step S1 based on embodiment 1.

[0080] In step S1, the three-dimensional coordinate point cloud data of the target area is acquired by the high-precision laser scanning radar carried by the UAV. The laser point cloud acquired by LiDAR is filtered and classified to establish a digital elevation model (DEM) of the target area.

[0081] Example 3

[0082] This embodiment further elaborates on step S2 based on embodiment 2.

[0083] In step S2, a computing payload developed by the developer is mounted on the UAV, and the position and attitude information of the UAV, including longitude, latitude, height, yaw, pitch and roll, is acquired at a frequency of 1Hz through interface programming.

[0084] Example 4

[0085] This embodiment further elaborates on step S3 based on embodiment 3.

[0086] In step S3, the real-time video stream from the drone camera is acquired through a computational processing load. The target recognition module built within the computational processing load processes and analyzes the real-time video stream, identifying potential threats to pipeline safety such as "buildings," "third-party construction," and "engineering vehicles," denoted as E. i Record Ei The coordinates of the top left corner (X) s ,Y s ) and the coordinates of the lower right corner (X) e ,Y e (and the corresponding timestamp.)

[0087] Example 5

[0088] This embodiment further elaborates on step S4 based on embodiment 4.

[0089] In step S4, the computational processing load aligns the key video frame data of the identified potential threat targets with the UAV pose data acquired in S2 using timestamps. If the video frame timestamps and pose timestamps are not perfectly aligned, the two sets of pose information closest to the video frame timestamps are used for interpolation. To interpolate and calculate the UAV position and attitude data at any given time point between two timestamps, linear interpolation or spherical linear interpolation is used.

[0090] S41: Linear interpolation calculates the interpolation values ​​for altitude and attitude angles (yaw, pitch, roll). For any attribute x, the formula for the difference between two time points t1 and t2 is:

[0091]

[0092] Where x(t) is the attribute value at time t, x(t1) and x(t2) are the attribute values ​​at two time points, t is the time point to be interpolated, t1 is the time point at which interpolation begins, and t2 is the time point at which interpolation ends.

[0093] S42: Spherical linear interpolation calculates longitude and latitude because these coordinates lie on the Earth's surface, which is a sphere.

[0094] △λ=λ2-λ1

[0095] a=sin((1-f)×Δλ)×sin(λ1)+sin(f×Δλ)×sin(λ2)

[0096] b=sin((1-f)×Δλ)×cos(λ1)+sin(f×Δλ)×cos(λ2)

[0097] λ(t) = atan2(a,b)

[0098] in,

[0099] λ1 and λ2 are the longitude or latitude of the start and end time points, respectively;

[0100] △λ is the difference between two coordinates;

[0101] f is the proportion from the starting time point to the interpolation time point, calculated as follows:

[0102] Example 6

[0103] This embodiment further elaborates on step S5 based on embodiment 5.

[0104] In step S5, the processing load is used to perform a projection transformation from camera coordinates to world coordinates on the key video frame, calculating the real spatial coordinate position corresponding to the video frame image and the projection of the pipeline onto the key video frame. The steps are as follows:

[0105] Establish the camera intrinsic parameter matrix K:

[0106]

[0107] c x =Image width / 2

[0108] c y =Image height / 2

[0109] Establish a rotation matrix R, and construct the rotation matrix R for each angle based on the yaw, pitch, and roll angles. z yaw), R y (pitch), R x (roll), the total rotation matrix is ​​the product of these matrices: R = R z (yaw)×R y (pitch)×R x (roll).

[0110]

[0111] Image coordinates to camera coordinates transformation: Given image coordinates (u, v), transform them to normalized camera coordinates (x, v). c ,y c The process is as follows:

[0112]

[0113] S51: Transformation from camera coordinates to world coordinates, camera coordinates (x... c ,y c 1) Transform to world coordinates (X) by inverse rotation and translation. w ,Y w Z w The ground elevation is obtained from a high-precision DEM pre-established in S1.

[0114] Ray direction:

[0115] Ray origin: o = R -1 ×(-T)

[0116] Intersection of the ray and the ground: P w =o+t×d

[0117] in:

[0118] t is the scaling factor of the ray from the starting point to the intersection point with the ground. The ray starts from the position of the camera and extends in a specific direction until it intersects with the ground.

[0119]

[0120] Z ground It is the ground elevation, obtained from the DEM created in S1.

[0121] o z It is the Z-coordinate of the ray origin (camera position) in world coordinates.

[0122] d z It is the component of the ray direction vector on the Z-axis.

[0123] T is the translation vector (T x ,T y ,T z ), that is, the latitude, longitude and altitude coordinates of the UAV obtained from S2.

[0124] S52: Transformation from world coordinates to camera coordinates, changing the coordinates of point P in the world coordinate system. w =[X w ,Y w Z w [Through rotation matrix R and translation vector T(T)] x ,T y ,T z Transform to camera coordinate system:

[0125] P c =R×P w +T

[0126] Projective transformation from camera coordinates to image coordinates, point P in the camera coordinate system c =[X c ,Y c Z c The image coordinates P are generated by projecting the camera intrinsic parameter matrix K onto the image plane. i =[u,v]:

[0127]

[0128] Using the steps of S51, potential threat events (X) s Y s), (X e Y e The coordinates are transformed to obtain the rectangular spatial range of the threat event in world coordinates.

[0129] Furthermore, spatial computation methods are employed to calculate E for each threat event. i The distance to the pipeline is used to determine the type of potential threat event. Based on the distance, it can be classified as encroachment, insufficient spacing, third-party construction, or no impact. The threat event is then labeled to obtain E. i ′.

[0130] Step S51 calculates the geospatial location of the entire image corresponding to the video keyframe. Spatial search analysis is performed by comparing the geospatial range corresponding to the image with the spatial coordinates of the actual pipeline route to find the pipeline route coordinates falling in the image. Then, the pipeline route is superimposed on the key video frame using the method in S52 to express the relative positional relationship between potential threat events and pipelines on the image.

[0131] Example 7

[0132] This embodiment further elaborates on step S6 based on embodiment 6.

[0133] E obtained in the S5 process i The results are displayed and alerted in GIS, and overlaid with the results of multiple rounds of identification to further determine which threat events are new and which are historically identified.

[0134] Example 8

[0135] An oil and gas pipeline threat event identification system is disclosed. The system consists of three parts: hardware, computing model, and data. The hardware includes a drone, a lidar, a camera, and a computing processing load. The computing model includes a digital elevation model, a position and attitude model, a pipeline threat target identification model, a data alignment model, a projection transformation model, and an overlay analysis model. The data includes hardware-acquired data and model analysis output data.

[0136] Hardware acquisition data refers to the vertical distance acquired by the lidar, the video stream and timestamp acquired by the camera, and the UAV flight control data, including longitude, latitude, height, yaw, pitch, roll, and coordinates of the threat target.

[0137] The embodiments of the present invention have been described in detail above, but the present invention is not limited to the described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention, and these equivalents or substitutions are all included within the scope defined by the claims of the present invention.

Claims

1. A method for identifying threat events in oil and gas pipelines, characterized in that, Includes the following steps: S1. Establish a digital elevation model of the target area; S2. Acquire the position and attitude data of the UAV; S3. Acquire real-time video streams from drone cameras and identify potential oil and gas pipeline threat events in the real-time video streams; S4. Align key video frame data of potential oil and gas pipeline threat events with UAV position and attitude data; S5. Using a digital elevation model, the camera coordinates of the aligned key video frames of the threat event are projected and transformed to world coordinates to calculate the real spatial coordinates of the threat event video frame image and the projection of the pipe in the key video frame. S6. Display the threat events.

2. The method for identifying oil and gas pipeline threat events as described in claim 1, characterized in that, Step S1 includes: acquiring three-dimensional coordinate point cloud data of the target area using a laser scanning radar mounted on a drone, filtering and classifying the three-dimensional coordinate point cloud data, and establishing a digital elevation model of the target area.

3. The method for identifying oil and gas pipeline threat events as described in claim 1, characterized in that, Step S2 includes: mounting a computing processing load on the UAV, wherein the computing processing load acquires the UAV's position and attitude information at a frequency of 1Hz through interface programming, wherein the position and attitude information includes longitude, latitude, altitude, yaw angle, pitch angle and roll angle.

4. The method for identifying oil and gas pipeline threat events as described in claim 1, characterized in that, Step S3 includes: acquiring real-time video streams from the drone camera using a computational processing load; processing and analyzing the real-time video stream using a pipeline threat target identification model built within the computational processing load; identifying potential pipeline security threats in the video stream; and denoting the identified targets as E. i Record E i The coordinates of the top left corner (X) s Y s ) and the coordinates of the lower right corner (X) e Y e (and the corresponding timestamp.) 5. The method for identifying oil and gas pipeline threat events as described in claim 1, characterized in that, Step S4 includes: The computational processing load aligns the key video frame data of the identified potential threat targets with the UAV pose data acquired in step S2 using timestamps; If the video frame timestamp and the pose timestamp cannot be perfectly aligned, then the two sets of data closest to the video frame timestamp in the pose information are used for interpolation.

6. The method for identifying oil and gas pipeline threat events as described in claim 5, characterized in that, In step S4, interpolation includes linear interpolation, which calculates the interpolation values ​​for height and attitude angle. For any attribute x, the attribute value between two time points t1 and t2 is: Where x(t) is the attribute value at time t, x(t1) and x(t2) are the attribute values ​​at two time points, t is the time point to be interpolated, t1 is the time point at which interpolation begins, and t2 is the time point at which interpolation ends.

7. The method for identifying oil and gas pipeline threat events as described in claim 5, characterized in that, In step S4, interpolation includes spherical linear interpolation, which calculates longitude and latitude, including: △λ=λ2-λ1 a=sin((1-f)×Δλ)×sin(λ1)+sin(f×Δλ)×sin(λ2) b=sin((1-f)×Δλ)×cos(λ1)+sin(f×Δλ)×cos(λ2) λ(t) = atan2(a,b) Where λ1 and λ2 are the longitude or latitude of the start and end time points, respectively; Δλ is the difference between the two coordinates; and f is the proportion from the start time point to the interpolation time point, calculated as follows: t is the time point to be interpolated, t1 and t2 are the time points being interpolated, a is an intermediate value calculated based on the longitude / latitude of the starting and ending points and the interpolation factor f, taking into account the changes in longitude / latitude and the sine value of the starting point's longitude / latitude, b is another intermediate value, combining the cosine values ​​of the starting and ending points' longitude / latitude and the sine value of the changes in longitude / latitude, and λ(t) is the longitude / latitude at time t.

8. The method for identifying oil and gas pipeline threat events as described in claim 1, characterized in that, Step S5 includes: using a camera coordinate to world coordinate conversion method to convert the coordinates of the potential threat event to obtain the rectangular spatial range of the threat event in world coordinates.

9. The method for identifying oil and gas pipeline threat events as described in claim 8, characterized in that, The S5 steps include: using spatial computation methods to calculate E for each threat event. i The distance to the pipeline is used to determine the type of potential threat event and to label the threat event.

10. The method for identifying oil and gas pipeline threat events as described in claim 1, characterized in that, The S5 steps include: By using a camera coordinate to world coordinate conversion method, the geographic spatial location corresponding to the entire image of the video key frame is calculated. Spatial search analysis is performed by comparing the geographic spatial range corresponding to the image with the spatial coordinates of the actual pipeline route to find the pipeline route coordinates falling in the image. By using the transformation method from world coordinates to camera coordinates, the pipeline route is superimposed with key video frames to express the relative positional relationship between potential threat events and pipelines on the image.

11. The method for identifying oil and gas pipeline threat events as described in claim 10, characterized in that, In step S5, the method for converting camera coordinates to world coordinates includes: converting camera coordinates (x, y, y) to world coordinates. c ,y c 1) Transform to world coordinates (X) by inverse rotation and translation. w ,Y w Z w ),in: Ray direction: Ray origin: o = R -1 ×(-T) Intersection of the ray and the ground: P w =o+t×d Where R is the rotation matrix and T is the translation vector (Ti). x ,T y ,T z ), that is, the latitude, longitude and altitude coordinates of the UAV obtained from step S2, t is the scaling factor of the ray from the starting point to the ground intersection point. The ray starts from the position of the camera and extends in a specific direction until it intersects the ground. Among them, Z ground It is the ground elevation, obtained from the digital elevation model established in step S1. z It is the Z-coordinate of the ray origin in world coordinates, d z It is the component of the ray direction vector on the Z-axis.

12. The method for identifying oil and gas pipeline threat events as described in claim 10, characterized in that, In step S5, the transformation method from world coordinates to camera coordinates includes: Point P in the world coordinate system w =[X w ,Y w Z w [Through rotation matrix R and translation vector T] (T x ,T y ,T z Transform to camera coordinate system P c =R×P w +T; Projective transformation from camera coordinates to image coordinates, point P in the camera coordinate system c =[X c ,Y c Z c The image coordinates P are generated by projecting the camera intrinsic parameter matrix K onto the image plane. i =[u,v]:

13. The method for identifying oil and gas pipeline threat events as described in claim 12, characterized in that, The camera intrinsic parameter matrix K is: c x =Image width / 2 c y =Image height / 2 The rotation matrix R is a rotation matrix R(yaw, pitch, roll) constructed from the rotation of each angle. z yaw), R y (pitch), R x (roll), the total rotation matrix is ​​the product of these matrices: R = R z (yaw)×R y (pitch)×R x (roll), where: Where roll represents the roll angle, pitch represents the pitch angle, and yaw represents the yaw angle.

14. The method for identifying oil and gas pipeline threat events as described in claim 1, characterized in that, The S6 steps include: displaying the threat events and overlaying them with the results of multiple rounds of identification to determine which threat events are new and which are historically identified.

15. An oil and gas pipeline threat event identification system based on the oil and gas pipeline threat event identification method according to any one of claims 1-14, characterized in that, include: Hardware modules include drones, LiDAR, cameras, and computing workloads; The computational models include digital elevation models, position and attitude models, pipeline threat target identification models, data alignment models, projection transformation models, and overlay analysis models; The data module includes hardware-acquired data and model-analyzed output data.

16. The oil and gas pipeline threat event identification system as described in claim 15, characterized in that, The hardware acquisition data includes vertical distance acquired by the lidar, video stream and timestamps acquired by the camera, and UAV flight control data such as longitude, latitude, altitude, yaw angle, pitch angle, roll angle, and coordinates of the threat target.

Images